File 019 19 months tracked 11 balance patches Score now 7.4, was 8.6
Patchline Weekly glyphPatchlineWeekly
Patch analysis · File 019

Verge Protocol Has Been Rebalanced Eleven Times. We Rebuilt Every Version to Find Out Which One Was Best.

Five weeks of playing nineteen months of history in order. The strongest build of this game shipped fourteen months ago, and each patch since has answered a complaint by removing something the game needed.

File opened 19 months ago11 balance patchesEvery build reconstructedScore now 7.4Originally 8.6

What we did

Verge Protocol has shipped eleven balance patches since release. Because the developer maintains downloadable historical builds for its own testing community, it is possible — unusually — to install and play each of those versions rather than reason about them from changelogs. Over five weeks we did exactly that, in release order, with the same four-person group playing the same modes on the same hardware.

This matters because the standard way to write about a live game's balance history is to read patch notes and interview players, both of which describe intention and memory rather than the game. We wanted the game.

The pattern

Eleven patches, and nine of them share a structure. A vocal section of the player base identifies something as frustrating. The next patch reduces or removes it. The frustration disappears, and so does a decision the player used to have to make.

Patch six removed the ammunition scarcity that made the mid-round retreat a real choice. Nobody complains about ammunition any more. Nobody retreats either, because there is no longer a reason to.

Taken individually each of these changes is defensible and several are clearly correct. Taken together, across nineteen months, they have converted a game about managing scarcity under pressure into a game about executing mechanically well, which is a thing many other titles already do better.

The four changes that cost the most

  • Ammunition scarcity, patch six. The most consequential single change. Removed the retreat decision entirely and shortened average round length by roughly a fifth in our recorded matches.
  • Respawn timing, patch eight. Halved, which eliminated the tension of holding a position with a depleted team. Rounds now recover from mistakes that used to decide them.
  • Map rotation cuts, patch nine. Three maps retired for being unpopular, all three of which rewarded patience rather than aggression. The surviving rotation is faster and more uniform.
  • Loadout unification, patch eleven. Class distinctions narrowed in the name of fairness, producing a roster where four of six roles now play very similarly.
A lone armoured figure standing in an empty concrete arena lit by overhead industrial lamps
One of the three maps retired in patch nine. It was unpopular and it was the most interesting space in the game.

Which build was best

Build 5, which shipped fourteen months ago and lasted eleven weeks. It had a genuine problem — one role was clearly overtuned in a way that dominated organised play — and it also had every structural element that has since been sanded away. Our group's assessment after replaying all eleven was unanimous, which surprised us more than it should have.

BuildRecorded average round lengthOur score at the timeScore on replay
Launch8m 40s8.68.4
Build 38m 20s8.68.5
Build 58m 05s8.48.8
Build 77m 10s8.07.9
Build 96m 35s7.67.5
Build 11 (current)6m 25s7.47.4

The part that is not the developer's fault

Every one of these patches was a reasonable response to feedback. The players complaining about ammunition scarcity were not wrong that it was frustrating; frustration was the mechanism. What the process cannot capture is that the people most likely to complain about a source of tension are the people least likely to want the game that tension produces, and there is no feedback channel through which the silent majority who quietly enjoyed a difficult decision can register that enjoyment.

This is a structural problem with iterating on a live competitive game, not a failure of judgement, and we suspect the same trajectory is visible in most titles that survive long enough to be patched eleven times. We are now looking for a counter-example in our other open files and have not yet found one.

Where the score stands

7.4, down from the 8.6 we originally published and from the 8.0 we revised to a year ago. This remains a competent, well-made competitive shooter with excellent movement and a technically clean build. It is also a considerably less interesting one than it was, and the specific quality that made it worth writing about at launch has been removed in eleven small, individually sensible steps.